Research map / Multi-agent systems and games / Cooperative multi-agent RL
CoMA-SLAM: Collaborative Multi-Agent Gaussian SLAM with Geometric Consistency
Lin Chen, Yongxin Su, Jvboxi Wang, Pengcheng Han, Zhenyu Xia, Shuhui Bu, Kun Li, Boni Hu, Shengqi Meng, Guangming Wang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| replace 3D Gaussians with 2D Gaussian surfels (ablation) | CoMA-SLAM · ReplicaMultiagent Office-0 | ATE RMSE (cm): 0.24 -> 0.10 | Improved |
| replace 3D Gaussians with 2D Gaussian surfels (ablation) | CoMA-SLAM · ReplicaMultiagent Office-0 | PSNR (dB): 42.21 -> 42.77 | No clear effect |
| add surface fitting initialization instead of random (ablation) | CoMA-SLAM · ReplicaMultiagent Office-0 | ATE RMSE (cm): 0.18 -> 0.10 | Improved |
| add surface fitting initialization instead of random (ablation) | CoMA-SLAM · ReplicaMultiagent Office-0 | PSNR (dB): 41.40 -> 42.77 | Improved |
| add loop closure (ablation) | CoMA-SLAM · ReplicaMultiagent Office-0 (per-agent eval) | ATE RMSE (cm): 0.21 -> 0.10 | Improved |
| add loop closure (ablation) | CoMA-SLAM · ReplicaMultiagent Office-0 | PSNR (dB): 42.73 -> 42.77 | No clear effect |
| add adaptive Gaussian growth (ablation) | CoMA-SLAM · ReplicaMultiagent Office-0 | Gaussian count (M): 1.05M -> 0.71M | Improved |
| add adaptive Gaussian growth (ablation) | CoMA-SLAM · ReplicaMultiagent Office-0 | ATE RMSE (cm): 0.10 -> 0.10 | No clear effect |
| use distributed design instead of centralized server | CoMA-SLAM vs MAGiC-SLAM · ReplicaMultiagent Office-0 | communication per frame: 1.7MB -> 2.2KB | Improved |
| use CoMA-SLAM instead of MAGiC-SLAM | multi-agent Gaussian SLAM · ReplicaMultiagent Apt-1 | ATE RMSE (cm): 0.26 -> 0.27 | No clear effect |
| use CoMA-SLAM instead of MAGiC-SLAM on tracking per-frame cost | multi-agent Gaussian SLAM · ReplicaMultiagent Office-0 | tracking time per frame (s): 0.10s -> 0.16s | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- MAGiC-SLAM — MAGiC-SLAM: Multi-Agent Gaussian Globally Consistent SLAM (CVPR 2025) · better on 8 of 11 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- MAGiC-SLAM: Multi-Agent Gaussian Globally Consistent SLAM (CVPR 2025 · similarity 0.91)
- MNE-SLAM: Multi-Agent Neural SLAM for Mobile Robots (CVPR 2025 · similarity 0.88)
- MAC-Ego3D: Multi-Agent Gaussian Consensus for Real-Time Collaborative Ego-Motion and Photorealistic 3D Reconstruction (CVPR 2025 · similarity 0.88)
- Gaussian Splatting SLAM (CVPR 2024 · similarity 0.87)
- DG-SLAM: Robust Dynamic Gaussian Splatting SLAM with Hybrid Pose Optimization (NeurIPS 2024 · similarity 0.87)
- GS-SLAM: Dense Visual SLAM with 3D Gaussian Splatting (CVPR 2024 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
